Claude Opus 5.5 · Anthropic

What does Claude Opus 5.5 cost to run in a coding agent?

Claude Opus 5.5 costs about $0.11 per average coding-agent request on our measured traffic, at list prices of $4 per million input and $20 per million output tokens. It is the strongest model for hard, long-running coding work: #1 on the Terminal-Bench 4.0 leaderboard at 64.9%.

List prices: Anthropic pricing, checked 2026-10-13. Request costs: Standard Compute gateway data, 79,301 coding-agent requests from 75 users, Sep 13 to Oct 12, 2026. Benchmark: Terminal-Bench 4.0 leaderboard. How we compute this

Run Claude Opus 5.5 in your agent from $19/month

See plans

What Opus 5.5 costs per coding-agent request

Cost at list price	Opus 5.5
Prompt under 100K tokens50% of requests, 87% of input cached	$0.053
Prompt of 100K to 272K39% of requests, 88% of input cached	$0.13
Prompt over 272K tokens11% of requests, 86% of input cached	$0.28
Average request	$0.11
100 requests	$10.66
Median session16 requests; a session ends after 10 idle minutes	$1.47
A month of regular usemedian of 20 users active on 5+ days	$155

Each logged request priced at the vendor's list rates for its own token counts, including cache reads and any higher rate for long prompts. Average output: 698 tokens per request.

What to watch for

Opus 5.5 costs exactly twice Sonnet 5.5 per request ($0.11 vs $0.053) because every Anthropic rate is doubled, while Sonnet 5.5 trails it by about three points on Terminal-Bench 4.0. Most sessions can run on Sonnet and switch to Opus for the hard steps. There is no long-prompt surcharge, but the 11% of requests over 272K tokens still make up 28% of its cost.

About Claude Opus 5.5

Claude Opus 5.5 is Anthropic's current flagship, released September 22, 2026. Anthropic says it performs at the level of Claude Fable 5.1 on most work while costing 40% less to run than Opus 5, and its docs recommend it as the default starting model. Independent boards agree it is at the top: it scores 58 on the Artificial Analysis Intelligence Index (max effort, #1 of the models tracked), leads the Terminal-Bench 4.0 leaderboard at 64.9% in Claude Code, and is #1 on LMArena's WebDev board. It has a 1M-token context window, 128K output, and always-on adaptive thinking with medium effort by default.

Sources: Anthropic announcement · Artificial Analysis · Terminal-Bench 4.0 · LMArena data

Claude Opus 5.5 FAQ

How much does Claude Opus 5.5 cost per coding-agent request?

$0.11 on average at list price, $10.66 per 100 requests, and $1.47 for a median 16-request session. That is 79,301 real coding-agent requests (Sep 13 to Oct 12, 2026) priced at Anthropic's list rates; the median request sent 99,534 input tokens, 87% of input was read from cache, and requests returned 698 output tokens on average.

Does Claude Opus 5.5 charge more for long prompts?

No. Claude Opus 5.5 bills the same rate per token up to its 1M-token context window.

Is Claude Opus 5.5 good for AI agents?

The hardest coding and agent work, as the default flagship instead of Fable. On our editorial scoring it rates 10.0/10 for agentic ability and 10.0/10 for output quality. #1 on the Artificial Analysis Intelligence Index (58) and Terminal-Bench 4.0 (64.9%).

What is Claude Opus 5.5's context window?

Claude Opus 5.5 has a 1M-token context window — large enough to hold sizeable codebases or document sets in a single run. Anthropic is the maker.

Can I self-host Claude Opus 5.5?

No — Claude Opus 5.5 is a proprietary, hosted-only model from Anthropic. You access it through an API (e.g. OpenRouter) rather than running the weights yourself.

What are the downsides of Claude Opus 5.5?

The main trade-offs: Twice Sonnet 5.5's price per token for a small measured gain; Released September 2026, so its long-run track record is still short. It's strongest for the hardest coding and agent work, as the default flagship instead of Fable.